
capture log close

set more off
set matsize 10000


*************************************************************************
*google trend time line from 2012 to 2019, change csv file to xlsx file.
*change <1 to 0.1

	clear
		import excel googletrend_timeline_from2012.xlsx, sheet("googletrend_timeline_from2012") clear

		*drop the first 3 observations which are words and variable names
		drop in 1/3

		*rename the variables
		encode A, generate(time) label(time)
		destring B, generate(gtindex)
		drop B

		
		tsset time
		graph twoway tsline gtindex, ///
					 ytitle("GTIndex") xtitle("Time") ///
					   xlabel(1 "2012.1" 13 "2013.1" 25 "2014.1" 37 "2015.1"  ///
							  49 "2016.1" 61 "2017.1" 73 "2018.1" 85 "2019.1", ///
							  labsize(vsmall)) 

		
		graph save figureA2.png, replace

